摘要
以天气雷达体积扫描的资料为基础,采用交叉相关法建立了一种对暴雨的临近预报方法。该方法的基本思路是:将第一时刻取得的雷达回波图像,向任一方向移过一定的距离,然后计算此图像与第二时刻图像之间的交叉相关系数,直到找到极大值,利用该极大值所对应的位移来预报下一时刻降水的位置。并利用“973”中国暴雨试验获得的观测资料对该方法进行了验证,验证结果表明:交叉相关法对暴雨有一定的预报能力,预报雨带的移动方向与雷达实际观测的移动方向相吻合,但其预报能力随着预报时间的增加而减弱。
Based on the Doppler radar data, a technique is described with the cross correlation method to heavy rainfall nowcast. Similar patterns of radar echoes are detected by comparing tracking areas in consecutive scans. The best fit between the tracking area is found by optimizing the correlation coefficient. The displacement vector corresponding with the max correlation coefficient is used to extrapolation rainstorm for periods between～quarter and one hour. The technique is validated by the use of observational data of ＂China Heavy Rain Experiment and Study CHeRES＂. It is showed that displacement directory of rainstorm nowcasted by cross correlation is consistent with that of radar observation, the forecast veracity is decreasing with forecast lead time.
出处
《气象》
CSCD
北大核心
2005年第10期12-15,F0003,共5页
Meteorological Monthly
关键词
交叉相关
临近预报
多普勒雷达
radar data cross correlation heavy rainfall nowcasting evaluation
